Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
FACULTY OF BUSINESS ::1 ·\: -. THE TWO NATIONS Of CANADA VS. THE NINE NATIONS OF NORTH AMERICA: A CROSS .. (UlTURAL ANALYSIS OF CONSUMERS' PERSONAL VALUES By Thomas E. Muller McMaster University Faculty of Business Hamilton, Ontario Marketing Area WORKING PAPER NO. 316 November? 7. 988 rER U�IVERSITY j eet West tario, Canada 6) 525-9140 L8S 4M4 The Two Nations of Canada vs . The Nine Nations of North America: A Cross-Cultural Analysis of Consumers' Personal Values Thomas E. Muller Faculty of Business McMaster University Hamilton, Ontario Canada November, 1988 This research was supported by Grant No. 494-85-0010 from the Social Sciences and Humanities Research Council of Canada. The Two Nations of Canada vs. The Nine Nations of North America: A Cross-Culturaf Analysis of Consumers' Personal Values Abstract Does the "Nine Nations of North America" scheme accurately reflect cross-subcultural differences among Canadians, or is the traditional French/English-Canada segmentation map more useful? This study of 1, 205 urban Canad ians living in three of the nine "nations"--Quebec, The Foundry, Ecotopia--measured their personal-value orientations in order to test the two segmentation schemes. The findings tentatively suggest that the French/English-Canada segmentation strategy is a better predictor of personal value differences than a three-nation approach. The growing interest in personal values f or cross-cultural analyses of consumer markets demands more research to refine the values measures. ' ,.; 'f1 2 In a recent empirical study, Kahle (1986) set out to test a concept that appear�d to have good prospects as a market segmentation tool based on regional dif terences in values and lifestyles--the Nine-Nations scheme for partitioning North Americans. The �arketing community on this continent was quite stirred up by Joel Garreau's (1981) book The Nine Nations of North America and his theory that North America's traditional boundaries were becoming meaningless as geographic segmentation schemes and as tools for predicting peoples' reactions to the events taking place around them. Garreau's thesis was that neither conventional political entities--e.g., the 12 Canadian provinces and territories or the 50 United States--nor standard regional designations (e.g., the West, the Prairies, Atlantic Canada, Central Canada or, in the U.S., the Midwest, West, South, Middle Atlantic) were valid or useful as a means of identifying differences in the values, loyalties, interests and lifestyles of North America's inhabitants. Rather, such differences were caused by numerous economic, social, cultural, topographical, and natural-resource f actors that defined "nations" with altogether different boundaries. It is these factors, Garreau argued, that shape the values of a nation's citizens, not the established political designations. Garreau's analysis--as participant-observer--of regional differences in values and lifestyles led him to posit that North America is actuall y a continent of nine distinct nations: Quebec, The Foundry, Ecot opia, New England, The Empty Quarter, The Breadbasket, Dixie, MexAmerica, and The Islands. A map of the nine nations reveals a North America that is radically different from the familiar political one (Garreau, 1981, pp. 204-205). For example, the Foundry, stretching along a north-south axis from Sudbury, Ontario to Cincinnati, Ohio and Washington, D.C., includes Chicago at its western border and New Haven, Connecticut at its eastern extreme. The nation 3 of Ecotopia is a narrow strip along the Pacific coast, extending from southern Alaska to just south of San Francisco, and includes Vancouverr Victoria, Seattle, and Portland. intact on his map. Quebec is the only "nation'' whose boundaries remain In terms of its citizens' values, Garreau claims, the Foundry city of Toronto, say, has a lot more in common with Detroit, Cleveland, Pittsburgh and Buffalo, than with other Canadian cities like Montreal in Quebec or Vancouver in Ecotopia. In sum, Garreau's Nine-Nations typology is based on his conjecture that regions with disparate physical environments, socio-political power, and economic prospects have evolved into nine distinct cultures in North America. These cultures have shaped its peoples' values which are reflected in their priorities, lifestyles, attitudes, outlooks on life, and consumption behavior, suggest ing that marketers have an alternate and, possibly, more relevant and timely method of segmenting North American consumers. Competing Value-Based Segmentation Maps From a Canadian marketing perspective, any new segmentation scheme would have to account for t he fundamental duality of a French and English Canada. True, there are many minority and ethnic subcultures among the Canadian people that represent opportunities for narrower segmentation. Stil l, the two founding cultural groups whose influence dates back to 1608, when Champlain founded Quebec, are numerically the two largest markets in Canada: 15. 7 million English Canadians and 6. 3 million French Canadians. This division of the country into two major cultures is recognized in many st udies that attempt to isolate differences in consumption behavior within Canada (see Kindra, Laroche & Muller, in press, chap. 8; Lawrence, Shapiro & Lalji, 1986; Schaninger, Bourgeois & Buss, 1985; Tamilia, 1980). The overwhelming majority of French Canadians reside in the province of 4 Quebec--by mother tongue, 81. 4% of Quebec's population of 6. 5 million is French-speaking, 8. 9% English-speaking {Statistics Canada, 1987). French Canada maintains a strong cultural identity stemming from religion, traditions and a dominant family orientation. French Canadians in Quebec have been depicted by their rural roots, minority status in Canada, North American orientation, Catholic mentality, Latin disposition and French outlook (Bouchard, 1980) . In contrast to the French Canadian, the English Canadian of Anglo-Saxon, Protestant background is pragmatic rather than theoretical, more social than individualistic, conforming rather than innovative, frugal rather than spendthrift, conservative rather than liberal (Henault, 1971) . contrast to the English Canadian, Again, in the French Canadian has an outlook that tends to be more humanistic, introspective, emotional, and less materialist ic (Kindra, Laroche & Muller, in press, chap. 8) . These kinds of differences suggest t hat the two cult ures are exerting a strong influence on their inhabitants that would be reflected in personal value differences. Thus, one way to evaluate Garreau's Nine-Nation typology from a Canadian position is to l ook for value-orientation profiles among Canadian consumers which parallel the "nation" of Quebec and other "nations" that include Canadian territory. The same can be done with French/English Canada as a segmentation plan, in order to determine which typology has the better ability to predict value profiles and is, therefore, of greater use to marketers. The purpose of the present study was to make this determination. Kahle's (1986) research pitted the Nine-Nation typology again� t U. S. Census Bureau regional designations in order to determine which geographic segmentation scheme better predicted differences in American consumers' personal values. Using data from a large-scale survey of mental health among American adults residing in the conterminous United States (Veroff, Douvan & 5 Kulka, 1981) , he matched a respondent's geographic location with that person's single most important value, as chosen from a list of nine terminal values. With respondents assigned to their respective "nations" as per Garreau's Nine Nation map, Kahle found that the proportion of people choosing various values did not change from nation to nation. On the other hand, the distribution of most important values among iesidents of a region did vary significantly across nine Census Bureau regions of the 48 conterminous United States. Kahle concluded that, as a value segmentation tool, the Census Bureau divisions were more useful than Garreau's Nine-Nations typology. While Kahle's study relied on a huge amount of data collected in face to-face interviews from a nationally representative U. S. sample, it remains incompl ete in two respects. First, it did not include a Canadian sample, as called for by the Nine-Nat ions boundaries, thus making his conclusions about no value differences across nations somewhat tent ative from a Canadian Second, people typically do not possess just one, all-important viewpoint. terminal value. As Rokeach (1973, p. 11) carefully points out, people have an organized system of values within which each acquired value is assigned a different priority. This total value system is stable enough to reflect an individual's distinct personality, yet sufficiently unstable to permit a reordering of value priorities when a person reacts to changes in society, culture and personal experience, as might occur, for instance, upon regional relocation a The present study adds to Kahle's (1986) contribut ion by 1. Obtaining Canadian data for the purpose of detecting cultural differences in value orientations, across two rival value segmentation schemes--Garreau's Nine-Nations typology and the traditional English/French-Canada typology; 6 2. Measuring several value orientations within a person's value system; Research on Consumer Values Values are cognitive,representations of various human needs, tempered by I societal demands (Rokeach, 1979, p. 48) . Values are guiding principles, and their relative import ance to a person reveal what the person deems worthwhile in life--what is worth striving for, achieving, fostering, supporting and protecting. People have terminal values, i. e. , desirable end-states of existence that could be labeled, for example, a sense of accomplishment, self -respect , security, an exciting life; people also have instrumental values--desirable modes of conduct conveyed by adjectives like ambitious, cheerful, courageous, honest, imaginative, polite, and so on, that enable people to achieve their t erminal values. People typically possess a system of values, or value hierarchy, within which individual values have been prioritized or assigned a relative importance. Hence, t he term value orientation denotes the importance attached to a generic value. Ten years ago, in the f irst edition of the Review of Marketing, an overview of the current consumer behavior literature indicated that research on values was still in an embryonic phase--in the introduction stage of the research-topic life cycle (Bettman, Kassarjian & Lutz, 1978) . The gradual gain in prominence of this topic in today's research literature hints that value research is approaching its growth stage (cf. Vinson, Scott & Lamont, 1977; Munson, 1984; Munson & McQuarrie, 1988) , though some observers believe ' that the topic has not received the emphasis it deserves (Cook, 1988, p. 8) . In either case, the proliferation of terms for the values concept.betrays its multidisciplinary nature and its continuing interest to marketing scholars. Various marketing studies refer to them as consumer values (Scott & Lamont, 7 197 3 ), cultural values (Henry, 1976), human values (Clawson & Vinson, 1918), personal values (Vinson & Munson, 1976; Munson & Mcintyre, 1978; Reynolds & Jolly, 1980; Pitts & Woodside, 1984), social values (Robertson, Zielinski & Ward, 1984, p. 3 54; Kahle, 1985), and psychological values (Garling, Lindberg & Montgomery, 1987). This scope of academic activity points to several promising areas of inquiry leading to a better grasp of the causes and effects of consumer values and where marketing pract itioners might put such knowledge t o work. A review of recent studies shows that empirical findings f rom value research can ultimately serve marketing managers, planners, and policy makers in at least five ways (see also Munson, 1984; Vinson, Scott & Lamont, 1977). One aspect of this research that should prove valuable is its application to cross-cultural market analysis. determined, Since values are culturally this focus should yield insights into a market's prevailing values and whether a contemplated marketing activity is compatible with core values. This applies to subcultures within a domestic market as well as to international marketing efforts. Though some headway has been made in this direction (Beatty, Homer & Kahle, 1988; Munson & Mcintyre, 1979; Powell & Valencia, 1984; Tse, Wong & Tan, 1988), the state of the art is still far short of a systematic, proven procedure for the cultural and value orientational analysis of a foreign market's potential. A second conceivably fruitful aspect is the academic work being done on the value measures, ·themselves. Several marketing scholars have turned to improving value measurement methodology and developing or adapting value invent ories that are relevant to consumption and feasible for survey research settings (Beatty et al., 1985; Kahle, Beatty & Homer, 1986; Munson & McQuarrie, 1988; Reynold� & Jolly, 1980). Standardized and validated procedures and instruments would be 8 immensely useful to both domestic and international marketers, and make findings and benchmarks much more comparable. A third aspect of academic research with potential applications is the identification of correlates and antecedents of consumers' value orientations. Only recently have marketing scholars begun to ident ify classes of variables that contribute to the devel opment and restructuring of peoples' value systems (e. g. , Crosby, Gill & Lee, 1984), and that could be used as clustering variables in value-segmentation strategies. Aside from culture and subculture, mentioned earlier, Munson (1984, p. 18} calls attention to both micro (e. g. , status in lifecycle, age, education, sex, IQ) and macro variables (e. g. , race, religion, reference groups) realignment of value priorities. that might govern the acquisition or If several of these prove to be powerful antecedents (or correlates) , then p ractitioners can achieve value segmentation simply by isolating clusters of consumers characterized by exact level s of just such micro/macro variables. A fourth promising aspect of this empirical work is the monitoring of personal value orientations in the population and the detection of societal value shifts. Demand forecasting, market planning, strategic management, and public policy depend not only on population projections in various segments, but also on information about value changes over time. The only way to obtain the latter is through periodic value surveys of national probability sam�les, similar in design to annual lifestyle surveys (e. g. , Horn & Wells, 1984). Though industry, not academics, took the initiative for monitoring values in the U.S. {Mitchell, 1983 ; Yankelovich, 1981), there are signs that academics will establish value monitors that are in the public domain (Kahle, Poulos & Sukhdial, 1988; Robin, 1984), although they are monstrously expensive and exact a long-term, programmatic research commitment. 9 Finally, a fifth aspect which has direct implications for marketing strategy is the impact of personal values on consumer preferences for products and brands. In search of value fulfillment, consumers seek out, identify with and consume products and services that can deliver attributes and ''consequences" related to specific personal values (Perkins & Reynolds, 1988; Reynolds & Gutman, 1984, 1988) . Similarly, certain everyday activities (Garling, Lindberg & Montgomery, 1987) and leisure/recreational alternatives (Pitts & Woodside, 1986) are cherished or pursued by consumers trying to fulf ill their value orientations. Thus, research findings which help identify segments possessing common value orientations allow the marketer t o design products and promotional strategies that remain consistent with a segment's values. To illustrate this application, consider the following examples from a study which used personal values to segment an international tourism market. Muller (1988) identified three principal segments of United States pleasure visitors to a Canadian city. Visitors in the largest of these segments placed a higher priority on the values security, a sense of belonging and being well respected (i. e. , being highly though t of by others) than tourists in the other segments. Such f indings imply that--to these people in particular--city attributes related to safety and security, such as low crime rates and clean, walkable streets are a strong attraction. It is important for them to feel welcome at a destination city; they want the feeling of familiarity and want to avoid the sense of being a stranger when they visit. Also important to members of this segment, in particular, is to avoid being embarrassed by cultural gaffes and looking foolish as foreign out-of-towners. Plenty of clear street signs, visitor information pamphlets, and guidance to points of interest in a city, would leave a greater impression on this segment than on others. Considering its value profile, this group offers the greatest r 10 potential for repeat-visit behavior, and marketing efforts to stimulate this demand should have a reassuring tone and build upon the destination' s familiarity and friendliness in these peoples' minds. Contrast this segmen� with a second one whose members are typically less preoccupied w�th �he values sec�rity, a sense of belonging, and being well respected. A visit to a foreign city by these tourists woul d not be dominated by the safet y-familiarity-friendliness considerations that are paramount to the first group. Relative to other segments, visitors in this second group place the lowest priority on the personal values self-respect, � relationships with others, . ,. a sense of accomplishment and self-fulfillment, and a greater importance on the values fun and enjoyment in life and excitement • Visitors in this segment are likely to be least concerned about establishing a rapport with local citizens and getting psychically or intellectually involved with the indigenous population. city. They do not mind being strangers in a foreign In fact, the strangeness of t he city--and t he anonymity it can offer-- probably add to the excitement of their experience. These people travel to a foreign destination to have fun, to let their hair down, to escape from the familiar. Sound ef forts t p develop tourism products for this segment will acknowledge these value orientations, then show how the foreign visit can spontaneously provide fun, enjoyment and sensory stimulation with a relatively low commitment or investment in effort, planning and familiarization on the visitor' s part. This marketing approach is very different from the one required for the first segment. ' In light of the foregoing literature review, the present study aims to cont ribute in the area of cross-cultural research on values and their application to segmentation. Based on the notion that personal values have cultural antecedents, this study evaluated two competing procedures for 11 segmenting urban Canadians by their value orientations. Canada, according to Garreau (1981} , is not one nation, but six different nations, three of which were the focus of this study--Quebec, The Foundry, and Ecotopia. Hence, do the personal values of urban Canadians differ across Quebec, The Foundry, and Ecotopia? This study tested t he proposition that the Garreau typology is a more useful means of segmenting the Canadian marketplace by geographic regions characterized by common values than the conventional, cultural division of English and French Canada. In keeping with the purpose of this study, the following general hypothesis was tested: H1: A three-nation (Quebec-The Foundry-Ecotopia} classification based on the Nine-Nation typology is a better predictor of value orientation segments among urban Canadians than the traditional French/English Canada segmentat ion scheme. Method Survey Design Residents of Canada's t hree largest metropolit an areas, Montreal, Toronto and Vancouver, were surveyed to represent Quebec, The Foundry, and Ecotopia, respectively. During April/May of 1987, a probability sample of 1, 205 adults (ea. 400 per city} was questioned via computer-assisted telephone interviewing from a central research facility in each city. Households were contacted by telephone, using an incremental random sampling technique which would include new listings and unlisted numbers in the sample frame. Within each household contacted, one respondent was selected randomly using a modification (Bryant, 1975} of the Troldahl and Carter (1964} procedure in order to balance sex and age among persons, 18 years and over, chosen for the 12 interview. Value Measures Respondents were asked to rate the importance they attached ("1" = "not important, " to "10" = "extremely important") to each of nine personal values: self-respect, security, warm relationships with others, a sense of accomplishment, fun and enjoyment in life, self-fulf illment, being well respected, a sense of belonging, excitement. This List of Values, a typology of terminal values (desirable end-states) developed at the University of Michigan Survey Research Center (Veroff, Douvan & Kulka, 1981, pp. 142-146), is based on the theoretic contributions of Maslow (1954), Rokeach (1973 ) and Feather (1975). Its psychometric properties are detailed in Kahle (1983 ). Research has shown (Beatty et al., 1985; Kahle, Beatty & Homer, 1986, p. 406) that the List of Values--which can be systematized to closely correspond with the human need categories in Maslow's hierarchy--relates more closely to major roles in life and everyday activities (work, leisure, marriage, parenting, consumption) than do the more widely known Rokeach (1973) Value Survey's 18 · terminal values that have also been employed in consumer research. The nine-item List of Values was professionally translated into French, for use among those Montreal respondents who preferred to be interviewed in French, rather than in English. interviewed in English, only. Citizens of Toronto and Vancouver were A reliability analysis was performed to check whether both English and French versions of the List of Values were internally consistent and free from random measurement error. The alpha coefficients (Cronbach, 1951) for the English (.80) and French (.76) versions indicate a relatively high reliability for both instruments (Churchill & Peter, 1984). Finally, to negate any systematic response error due to order, the presentation order of the values was rotated f rom interview to interview. 13 Results Table 1 shows the mean importance rating given to nine personal values by citizens of Quebec, The Foundry and Ecotopia. more important that value is to residents. The higher the mean, the To enhance their interpretation, the nine values have been blocked and ordered in the table to correspond with the five human need categories in Maslow's (1954) hierarchy (see Veroff , Douvan & Kulka, 1981, pp. 144-145). Thus, excitement and fun and enjoyment in life are values expressive of physiological needs; security is the value driven by safety needs; a sense of belonging and warm relationships with others are expressive of belongingness and love needs, and so forth, up the hierarchy. Though one-way ANOVAs reveal that value orientations differ, overall, at � < .05 for four values and � < .10 for two other values, an interesting pattern is revealed by multiple-comparison tests of differences at � � .05. In six out of nine cases, it is Quebec that differs significantly from either Ecotopia or The Foundry. In one case, only, (security) do The Foundry and Ecotopia differ from each other in value orientation. To check how well this set of value orient ations predicts "nationhood, " the data were submitted to a three-group discriminant analysis. sample was randomly split in half. The entire One half (602 cases) was used as an analysis sample to estimate the t wo discriminant functions. Then, the functions' ability to correctly classify citizens of the three nations was tested on the hold-out sample. Only Q.!1g_ of the two functions was statistically significant and it accounted for virtually all of the variance attributable to differences among the groups (Wilks' lambda 51.9, � < .001). = .92; The classification results appear in Table 2. ::( 2 (18) = Overall, the three-nat ion scheme results in 42.5% correct classifications (an improvement of 9.2% over the results expected by chance) and the proportion correctly 14 classified varies from 20% for The Foundry, to 64% for Ecotopia. Note that the majority of Foundrians have been misclassified as Ecotopians, even though the prior odds of being correctly classified are virtually the same for each nation (�.33). Table 2. A clearer picture is obtained from the last two columns in Of the 184 persons classified as Quebecois, 89 (or 48.4%) were actually Quebecois. 33.7. That is a 14.7% improvement over the chance percentage of However, t he predictions for The Foundry and Ecotopia are worse, representing improvements over chance of 5.4 and 7.6, respectively. Table 1 and Table 2 about here The findings, thus far, do not bode well for the Three-Nation conception of urban Canada. Tables 1 and 2 imply, however, that Ecotopia and The Foundry are much closer to each other in personal value orientations--and that both differ from Quebec. Since the proportion of French Canadians in the populations of Toronto (The Foundry) and Vancouver (Ecotopia) are 1.5% and 1. 6%, respectively, and the two cities are overwhelmingly English Canadian (Statistics Canada, 1984), the observed differences suggested a Quebec vs. English Canada dichotomy, i.e., more in alignment with the t raditional French/ English-Canada division. In Montreal (Quebec), however, the proportion of English Canadians (at least by mother tongue) is about 18%--vs. 70% French Canadians. To correctly test the segmentation scheme of French vs. English Canada, the Quebec sample was partitioned into 320 French Canadians and 85 English Canadians, and the latter were pooled with the citizens of Toronto and Vancouver to represent a sample of 885 English Canadians. Research done in Metropolitan Montreal has shown that the measure, language of returned bilingual mail questionnaire, correctly identifies French 15 and English Canadians in 90.4% of cases (Bergier, 1986). In the present study, Montrealers were given the option of being interviewed in French or English. Thus, language of interview was the basis for separating Montrealers into French and English Canadians. Table 3 gives the results of a random, split-sample discriminant analysis of French vs. English Canada, where half the sample was used to derive the discriminant function and the remaining cases subsequently served as the hold-out validation sample. Correlations between personal-value importance scores and overall discriminant scores are also shown. These indicate which personal values best differentiate between French and English Canadians. The mean importance scores in the last two columns show in which direction the two cultures differ in their value orientations. In contrast to their anglophone countrymen, French Canadians collectively come across as more security minded, and more concerned about having the respect of others. Also, French Canadians put a lower priority on the values a sense of accomplishment, fun and enjoyment in life, and excitement. These results seem to underscore the strong French-Canadian orientation toward tradition, religion, the central role of the family unit in their lives, their rural roots and, historically, their minority status in Canada. The lower priority placed on a sense of belonging would support earlier research findings that French Canadians are more individualistic than English Canadians. The more pragmatic, conformist, establishmentarian and socially oriented (less individualistic) English Canadian is revealed through the higher priority that this cultural group gives to the values a sense of belonging, a sense of accomplishment, and values that are expressive of physiological needs, i.e., fun, enjoyment, excitement. Table 4 shows how the estimated discriminant function classified the 16 hold-out sample. The earlier, Three-Nation discriminant functions had correctly classified 43 % of citizens by their value orientations. The single, two-group function was able t o correctly predict 72% of French and English Canadians. Evaluating these results with the appropriate chance models in mind, the overall rate of 72% correct classifications is a 12% improvement on chance (Cpro. = 60%) . Even though the correct prediction of French Canadians appears dismally low, one should remember that classifying a person as a French Canadian defies the prior odds, since only 28% of the sample (and, incidentally, about 24% of Canada's population) are actually French Canadian. The performance of this two-group discriminant function is most clearly seen in the last two columns. Of the 43 people classed as French Canadian, 21 or 48.8% were correct determinations--a 20.7% improvement on chance. these findings do not support H1. overall, Contrary to the hypothesis, the French/English-Canada dichotomy is a somewhat better predictor of value orientations than a Three-Nat ion typology based on Garreau's Nine Nations. Moreover, the specific value orientations that differentiate French and English Canadians are in line with findings reported in the literature dealing with these two major cultural groups in Canada. Table 3 and Table 4 about here Discussion With the increased interest in personal values as a potent variable in marketing strategies aimed at consumers, the question arises as to how consumers can be effectively segmented by their value orientations. Garreau's (1981) Nine-Nations concept has been attractive for its possibility as one such segmentation tool. Applied to the Canadian marketplace, six of these 17 "nations" would be relevant, of which three were the focus of t his study. Since culture is one of the determinants of values, an alternate value segmentation tool is the French/English-Canada dichotomy that has long been in use among marketers catering to both cultural groups. A major purpose of this study was to determine which of two value-segmentat ion schemes was· more valid and useful in the Canadian context. The findings suggest t hat differences in consumers' value orientations are more apparent when one divides the population into French and English Canadians, rather than into Ecotopians, Foundrians and Quebecois, as per the Nine-Nations scheme. The findings also imply that marketers serving both French and English Canadians can continue to rely on this traditional cultural dichotomy in order to devise a successful marketing mix and avoid making marketing mistakes. These data support Kahle's (1986) findings when the same List of Values was applied to the United States population, segmented according to Garreau's map of the nine nations. The Nine-Nations typology failed to differentiate people by their principal value orientation. On t he other hand, these conclusions are tempered by the fact that the French/English-Canada cultural dichotomy does not explain all of the differences in value orientations among Canadians. Conversely, by relying on value orientations, alone, many French Canadians were misclassified as English Canadians, and the overall rate of correct predictions of cultural group membership is only 72%. Thus, from a practical standpoint, personal values cannot account for all the differences that separate English Canadians from French Canadians as distinct marketing entities. What this implies is that marketers serving both cultural groups in Canada will need to base their product design and promotional strategies on more than just information about personal value orientations. Of course, we need to recognize that factors 18 other than culture will influence a person's value orientations. Since a consumer's values stem from psychological needs that have been tempered by societal demands (Rokeach, 1973 ), culture is but one antecedent of personal values . This suggests an alternate scheme for segmenting consumer markets in Canada: by clusters of consumers with distinctive personal-value profiles, irrespective of, or in addition to, geographic location and cultural group membership. To date, research done on personal values by marketing academics indicates that value-segmentation can lead to valuable insights about consumption motives and patterns in the marketplace. Study Limitations In all fairness to Garreau's Nine-Nations-of-North-America theory, given the attention it has generated in U. S. and Canadian marketing circles (e. g. , Major, 1983 ; Vredenburg & Thirkell, 1983 ; Whalen, 1983 ), this study represents a partial test of his regional typology among Canadians. First, it is based entirely on value orientations, and on only three of the six nations. The "nations" of New England, The Breadbasket and The Empty Quarter, making up the remainder of Canadian territory, were not examined for differences in values. Second, the conclusions about value differences are limited to populations living in major urban areas in the three nations examined. Only citizens of Montreal (Quebec), Toronto (Ontario} and Vancouver (British Columbia) were sampled--although it should be noted that about 80% of the populations of Quebec, Ontario and British Columbia live in urban centers. A more comprehensive test of the theory's utility as a value-based segmentation tool in Canada would include probability samples from both rural and urban areas in all six nations. Differences among nations would then be tested against alternate regional designations that have been used in Canada� Another limitation of the study is its aggregation of Toronto and 19 Vancouver citizens into one sample representing English Canada. It is acknowledged that any detailed study of regional value orientations would have to isolate the relatively small but real presence of French Canadians outside Quebec, as well as ethnic minorities in each region {e.g., Italian, German, Chinese Canadians) Canadians. who may not possess the value orientations of English Additional refinements in research of this type should also include an examination of differences between Canadians of British origin and Canadians whose mother tongue is listed as English by Statistics Canada. Future Research From t he viewpoint of marketing in Canada, future studies of consumer values will need to ponder the ethnic diversit y of Canadians and the possibility t hat value-based segmentation can become more useful if fine-tuned to these differences. Unlike the United States, which is de�icted as a melting pot that tends to assimilate ethnic groups, Canada is often described as a salad bowl whose ingredients remain relatively distinct while mixing with each other. Some of this distinctness will likely show up in culturally unique value orientations with consequent economic behaviors. From the viewpoint of value research methodology, a study of this nature underscores some of t he areas that need further work. which value inventory to use in market studies. One is the issue of The three most commonly used inventories have been the Rokeach Value Survey, the proprietary VALS {Values and Life Styles) items, and the newer and much shorter List of Values. Though some marketing researchers have proposed the need for very specific consumption-oriented value inventories, only further research can tell which approach yields value profiles that are reliable, complete and actionable for the marketer. Another area is the actual measurement of value orientations, i.e. , 20 getting a grasp on how a given set of values have been prioritized by the consumer. In this study, the 10-point Likert scale with which value importance measures were taken over the telephone tended to elicit importance ratings at the high end of the scale. Apparently, people are reluctant to assign a low importance to what are obviously supposed to be important things in life to many people. Research focusing on personal value measurement has shown that this "end-piling phenomenon'' is not serious and therefore does not outweigh the usefulness of the rating-scale approach in marketing applications (Munson & Mcintyre, 1979) . However, if, as Rokeach (1973) points out, people distinctly align values in some sort of hierarchy of importance to them, then end-piling works against a clearly articulated system of values. A creative breakthrough for teasing out real (and at least interval-scaled) differences in value priorities, in a manner that is feasible in telephone and mail survey research, would be a very welcome contribution research. to value-based marketing 21 References Beatty, S.E., Homer, P.M. and Kahle, L.R. (1988) . Problems with VALS in International Marketing Research: An Example from an Application of the Empirical Mirror Technique. In M.J. Houston (Ed.), Advances in Consumer Research {Vol. 15, pp. 375-380) , Provo, UT: Association for Consumer Research. Beatty, S.E., Kahle, L.R., Homer, P.M. and Misra, s. {1985) . Alternative Measurement Approaches to Consumer Values: The List of Values and the Rokeach Value Survey, Psychology and Marketing, 2 (Fall), 181-200. Bergier, M.J. (1986) . Predictive Validity of Ethnic Identification Measures: An Illustration of the English/French Classification Dilemma in Canada, Journal of the Academy of Marketing Science, 14 (Summer), 37-42. Bettman, J.R., Kassarjian, H.H. and Lutz, R.J. (1978) . Consumer Behavior. In G. Zaltman and T.V. Bonoma (Eds.), Review of Marketing 1978 (pp. 194-229) , Chicago: American Marketing Association. Bouchard, J. (1980) . Differences, Montreal: Editions Heritage. Bryant, B.E. (1975) . Respondent Selection in a Time of Changing Household Composition, Journal of Marketing Research, 12 {May), 129-135. Churchill, G.A., Jr. and Peter, J.P.· {1984) . Research Design Effects on the Reliability of Rating Scales: A Meta Analysis, Journal of Marketing Research, 21 {November), 360-375. Clawson, C.J. and Vinson, D.E. {1978) . Human Values: A Historical and Interdisciplinary Analysis. In H.K. Hunt {Ed.), Advances in Consumer Research {Vol. 5, pp. 396-402) , Ann Arbor, MI: Association for Consumer Research. Cook, W.A . {1988) . On Values and the Valued [editorial], Journal of Advertising Research, 28 {February/March), 7-8. 22 Cronbach, L.J. (1951). Coefficient Alpha and the Internal Structure of Tests, Psychometrika, 16 (September) , 297-3 3 4. Crosby, L.A., Gill, J.D. and Lee, R.E. (1984). Life Status and Age as Predictors of Value Orientation. In R.E. Pitts, Jr. and A.G. Woodside (Eds.) Personal Values and Consumer Psychology (pp. 201-218), Lexington, MA: Lexington Books. Feather, N.T. (1975} . Values in Education and Society, New York: Free Press. G�rling, T., Lindberg, E. and Montgomery, H. (1987). Beliefs About Attainment of Lif e Satisfaction as Determinants of Preferences for Everyday Activities. In Understanding Economic Behaviour (Vol. 3 , pp. 821-83 0), Proceedings of the 12th Annual Colloquium of the International Association for Research in Economic Psychology. Aarhus, Denmark: Handelshojskolen. Garreau, J. {1981) . The Nine Nations of North America, New York: Avon Books. Henault, G. {1971). Les consequences du biculturalisme sur la consommation, Commerce, 73 (September) , 78-80. Henry, W. A. (1976). Cultural Values Do Correlate With Consumer Behavior, Journal.of Marketing Research, 13 (May) , 121-127. Horn, M. I. and Wells, W.D. (1984). Do Trends in Attitudes Predict Trends in Behavior? In R.E. Pitts, Jr. and A.G. Woodside (Eds.) Personal Values and Consumer Psychology {pp. 187-198), Lexington, MA: Lexington Books. Kahle, L.R. (Ed.) (1983 ). Social Values and Social Change: Adaptation to Life in America, New York: Praeger. Kahle, L.R. {1985). Social Values in the Eighties: A Special Issue, Psychology & Marketing, 2 (Winter) , 23 1-23 7. Kahle, L.R. (1986). The Nine Nations of North America and the Value Basis of Geographic Segmentation, Journal of Marketing, 50 {April) , 3 7-47. Kahle, L.R., Beatty, S.E. and Homer, P. (1986). Alternative Measurement 23 Approaches to Consumer Values: The List of Values (LOV) and Values and Life Style (VALS) , Journal of Consumer Research, 13 (December) , 405-409. Kahle, L.R., Poulos, B. and Sukhdial, A. (1988) . Changes in Social Values in the United States During the Past Decade, Journal of Advertising Research, 28 (February/March) , 35-41. Kindra, G. S., Laroche, M. and Muller, T.E. (in press) . Consumer Behaviour in Canada, Toronto: Nelson Canada. Lawrence, C., Shapiro, S.J. and Lalji, S. (1986) . Ethnic Markets: A Canadian Perspective, Journal of t he Academy of Marketing Science, (Summer) , 7-16. Major, M.J. (1983, September 6) . Ecotopia: Bellwet her Region Offers Look at Future Life Styles, Marketing News, p. 6. Maslow, A.H. (1954) . Motivation and Personality, New York: Harper. Mitchell, A. (1983) . The Nine American Life Styles, New York: Warner. Muller, T.E. (1988) . The Role of Personal Values in Segmenting an International Tourism Market (Working Paper No. 313) , Hamilton, Ontario: McMaster University, Faculty of Business. Munson, J.M. (1984) . Personal Values: Considerations on Their Measurement and Application to Five Areas of Research Inquiry. In R.E. Pitts, Jr. and A.G. Woodside (Eds.) Personal Values and Consumer Psychology (pp. 13-33) , Lexington, MA: Lexington Books. Munson, J.M. and Mcintyre, S.H. (1978) . Personal Values: A Cross Cultural Assessment of Self Values and Val ues Attributed to a Distant Cultural Stereotype. In H.K. Hunt (Ed.) Advances in Consumer Research (Vol. 5, pp. 160-166) , Ann Arbor, MI: Association for Consumer Research. Munson, J. M. and Mcintyre, S.H. (1979) . Developing Practical Procedures for the Measurement of Personal Values in Cross-Cultural Marketing, Journal of Marketing Research, 16 (February) , 48-52. 24 Munson, J .M . and McQuarrie, E .F. (1988). Shortening the Rokeach Value Survey for Use in Consumer Research . In M .J. Houston (Ed.) Advances in Consumer Research (Vol. 15, pp. 381-386), Provo, UT: Association for Consumer Research. Perkins, W .S. and Reynolds, T .J . (1988). The Explanat ory Power of Values in Preference Judgements: Validation of the Means-End Perspective. In M.J. Houston (Ed .) Advances in Consumer Research (Vol. 15, pp. 122-126), Provo, UT: Association for Consumer Research . Pitts, R .E., Jr. and A .G. Woodside (Eds.) (1984). Personal Values and Consum.er Psychology, Lexington, MA: Lexington Books . Pitts, R.E., Jr . and A .G. Woodside (1986). Personal Values and Travel Decisions, Journal of Travel Research, 25 (Summer), 20-25. Powell, T .E. and Valencia, H. (1984) . An Examination of Hispanic Subcultural and Regional Value Orientations. In R.E . Pitts, Jr . and A.G. Woodside (Eds.) Personal Values and Consumer Psychology (pp . 219-230}, Lexington, MA: Lexington Books . Reynolds, T.J . and Gutman, J . (1984). Laddering: Extending the Repertory Grid Methodology to Construct Attribute-Consequence-Value Hierarchies . In R.E. Pitts, Jr . and A .G . Woodside (Eds .) Personal Values and Consumer Psychology (pp. 155-167), Lexington, MA: Lexington Books . Reynolds, T.J . and Gutman, J. (1988). Laddering Theory, Method, Analysis, and Interpretation, Journal of Advertising Research, 28 (February/March), 11-31. Reynolds, T .J. and James P . Jolly (1980). Measuring Personal Values: An Evaluation of Alternative Methods, Journal of Marketing Research, 17 (November), 531-536. Robertson, T.S., Zielinski, J. and Ward, S. (1984). Consumer Behavior, Glenview, IL: Scott, Foresman and Company. 25 Robin, D.P. (1984} . The Logic of Establishing Value Monitors. In R.E. Pitts, Jr. and A.G. Woodside (Eds.) Personal Values and Consumer Psychology (pp. 71-75} , Lexington, MA: Lexington Books. Rokeach, M. (1973 } . The Nature of Human Values, New York: Free Press. Rokeach, M. (1979). From Individual to Institutional Values: With Special Reference to the Values of Science. In M. Rokeach (Ed.) Understanding Human Values: Individual and Societal (pp. 47-70} , New York: Free Press. Schaninger, C.M., Bourgeois, J. and Buss, W.C. (1985). French-English Canadian Subcultural Consumption Differences, Journal of Marketing, 49 (Spring} , 82-92. Scott, J.E. and Lamont, L.M. (1973 ). Relating Consumer Values to Consumer Behavior: A Model and Method for Investigation. In T.W. Greer {Ed.} Increasing Marketing Productivity {pp. 283 -288} , Chicago: American Marketing Association. Statistics Canada (1984). Metropolitan Atlas Series (1981 Census of Canada, Catalogues 99-919, 99-920, 99-921, July, pp. 26-28), Ottawa: Minister of Supply and Services Canada. Stat istics Canada (1987). 1986 Census of Canada, Summary Tabulation (Table MT86A01A, July), Ottawa: Minister of Supply and Services Canada. Tamilia, R.D. (1980). Cultural Market Segmentation in a Bilingual and Bicultural Setting, European Journal of Marketing, 14 (4), 223 -23 1. Troldahl, V.C. and Carter, R. E. (1964). Random Selection of Respondents Within Households in Phone Surveys, Journal of Marketing Research, 1 (May} , 71-76. Tse, D.K., Wong, J.K. and Tan, C.T. (1988). Towards Some Standardized Cross Cultural Consumption Values. In M.J. Houst on (Ed.} Advances in Consumer Research (Vol. 15, pp. 3 87-3 95), Provo, UT: Association for Consumer Research. 26 Veroff, J., Douvan, E. and Kulka, R.A. (1981). The Inner American: A Self Portrait from 1957 to 1976, New York: Basic Books. Vinson, D. E. and Munson, J.M. (1976). Personal Values: An Approach to Market Segmentation. In K.L. Bernhardt (Ed.) Marketing: 1776-1976 and Beyond (pp. 313-317), Chicago: American Marketing Associat ion. Vinson, D.E., Scott, J.E. and Lamont, L.M. (1977). The Role of Personal Values in Marketing and Consumer Behavior, Journal of Market ing, 41 (April) , 44-50. Vredenburg, H. and Thirkell, P. (1983). Canadian Regionalism: A Marketplace Reality? In J.D. Forbes (Ed.) Marketing 1983: Linking Knowledge to Action (pp. 360-370). Proceedings of the Annual Conference of the Administrative Sciences Association of Canada--Marketing Division, Vol. 4, Vancouver, BC: Administrative Sciences Association of Canada. Whalen, B. (1983, January 21). The Nine Nations of North· America: 'Provocat ive' Theory Gives Researchers a New View of Consumer, Market Differences, Marketing News, pp. 1, 19. Yankelovich, D. (1981). New Rules: Searching for Self-Fulfillment in a World Turned Upside Down, New York: Bantam. 27 Table 1 Mean Value-Importance scores for Three Nations and One-Way ANOVA Results Significance Personal values (blocked & ordered to correspond with Maslow's need hierarchy) QUEBEC (� = 405) THE FOUNDRY (� = 3 98) ECOTOPIA (� = 402) of overall differences8 (Self-actualization) ns Self-fulfillment 8.40 8.3 5 8.41 A sense of accomplishment 8.14A 8.3 0 8.42A .0754 Being well respected 8.53 AB 8.23 8 8.00A .0001 Self-respect 9.06 9.08 8.93 A sense of belonging 7.3 3 8 7.668 7.54 Warm relationships w/others 8. 42 8.43 8.3 3 8.64A 8.42C 8.19AC .0006 Fun and enjoyment in life 7.84AB 8.118 8.20A .0128 Excitement 6.0SAB 6. 758 6.64A .0000 (Esteem) ns (Belongingness and love) .0699 ns (Safety) Security (Physiological) 8Group means with identical superscripts differ significantly at � � .05 in Duncan multiple-range tests. 28 Table 2 Classification Results for Hold-Out Sample from Three-Group Discriminant Analysis Predicted group Correct No. of cases predictions No. of cases (Percent of actual) (% of Actual group QUEBEC THE FOUNDRY ECOTOPIA Totals THE total) QUEBEC 203 89 31 83 (33. 7) (43. 8) (15.3) {40.9) 202 57 41 104. (33. 6) (28. 2) (20.3) (51.5) 197 38 33 126 (32. 7) (19.3) (16. 8) (64.0) 602 184 105 313 FOUNDRY Proportional chance criterion, Cpro., = ECOTOPIA Percent improve- as a % ment of total over predicted chance 48.4 14.7 39.0 5. 4 40.3 7. 6 .333 Overall rate of correct classifications: 42.5% 9.2 29 Table 3 Discriminating Power of Personal Values in a Two-Group Discriminant Analysis and Differences in Value Orientations Between French and English Canadians Mean (SD) Correlation between value and discriminant Personal valuea functionb Excitement value-importance ratingc French Canada (n, = 320) English Canada <n. = 885) .57 5.87 (2.53) 6.70 (2.15)*** 42 7.79 (1.97) 8.14 (1. 70)** -.36 8.-49 {1.67) 8.17 (1.83)** .33 8.10 (2.00) 8.35 (1.58}* -.30 8.61 (1.74) 8.35 (1.66)* .26 7.16 {2.27) 7.64 (2.00)*** Warm relationships with others -.13 8.41 (1.83) 8.38 (1.69) Self-respect -.07 9.01 ( 1.46) 9.03 (1.35) .04 8.44 (1.71) 8.37 (1.57) Fun and enjoyment in life . • Being well respected A sense of accomplishment Security A sense of belonging Self-fulfillment aordered in descending discriminating power. hBased on the 602 cases in the split sample used for analysis, but not for classification. cTwo-tailed 1 tests reveal that differences in means are significant at ***� � .001, **� � .005, *� < .05. 30 Table 4 Classification Results for Hold-Out Sample from Two-Group (French/English Canada) Discriminant Analysis Predicted group No. of Actual group French Canada English Canada English of total over Canada predicted (% of French total) Canada 169 21 (28.1) (12.4) (87.6) 22 411 602 148 (5 .1) (94.9) 43 559 Proportional chance criterion, Cpro., = improvement (Percent of actual) 433 predictions Percent as a % cases (71.9) Totals No. of cases Correct chance 48.8 20.7 73.5 1.6 .596 Overall rate of correct classifications: 71.8% 12.2 Faculty of Bus ine s s McMaster Univers i ty WORKING PAPERS 295 . J ohn W . Medc o f , Que s tionnaire : 296 . J ohn W . - RECENT RELEAS E S " Ne e d for Dominance S c al e o f the Manife s t Needs I t s Rel i ab i li ty and Val i d i ty " , Medcof and Michael J . S tahl , A T e s t o f Chusmir ' s Model " , Apr i l , 297 . Zvi Golds te in , Shaul P . Apr i l , 1988 . " J ob - Mo tive Comp a t ib il i ty : 1988 . Ladany and Abraham Mehrez , "A Replacement Model W i th An Exp e c t e d Breakthrough in an Improving Techno l o g i c al Environment " , 298 . Apr il , Norman Arche r , 1988 . "A Compar i s on o f C omputer Conferences w i th Face - to Face Mee t ings for Small Group Bus ine s s D e c i s ions " , 299 . J o s eph Ro s e , June , 300 . May , 19 8 8 . " New Measures o f Union Organizing Effe c t ivenes s " , 1988 . Min Bas adur , "App l ication of a Comp l e t e Pro c e s s o f App l ie d Creative Problem S o lving to Union-Management Bargaining " , June , 301 . Rick D . June , 302 . 1988 . A. Hacke tt , 19 8 8 . "Age , Tenure , and Emp l oyee Ab s ente e ism " , Will i am R i chardson and Abraham Mehrez , S c al e Mul t i - Echelon S ervice Sys tems " , I" 303 . Min Basadur , 11 Inve s t igation o f Large June , 1988 . " The New Creative Thinking S k i l l s Todays Purchas ing Profe s s ionals Mus t Have to Be Suc c e s s ful " , June 1 9 8 8 . 304 . George S te iner , " On the Complex i ty o f Dynamic Pro gramming for S equencing Prob lems with Precedence Cons traint s " , 305 . I s ik Zeytino g l u , Augus t , 306 . Tom Mul l e r , Roy J . Trends 308 . Roy J . Gary N . in Turkey ' s Lab our Marke t " , "Value Orientations in Urban Canadians Acro s s the Span " , Adams , July , 1988 . " North Ame r i c an Indus trial Re lations : in Canada and the United S tates " , Adams , " The Comp e t i t ivene s s : 309 . 19 8 8 . 1988 . Adult Life 307 . " Gender D ivis ions July , Augus t , " Old Indus trial Re lat ions " A Canadian Cas e " , Cha i s on and Jos eph B . Ro s e , Augus t , D ivergent 1988 . and C o rp o r ate 1988 " Continental D ivide : D ir e c t ion and Fate o f No rth Ame r i c an Unions " , Augus t , The 1988 . 2 310 . Thomas E. Mul l er , "Us ing Consume r s ' the Canadi an Marketplace " , 311 . Yufe i Yuan , of B ias Value Orientati ons Sep temb e r , Abraham Mehrez , Amiram Gafni , in a Pers onnel As s ignment Problem : Mul tip l ic a t ive U t i l i ty S o luti�ms " , Yufe i Yuan and Abraham Mehrez , S e gment "A S ta t i s t i c a l Analys i s Octob e r , S tab le S o luti ons Vs . ... 1 9 8 8" f 3 12 . to 1988 . " Re fuel ing ·s trate g i e s , . • ·� \.;.. The Operat i onal Range With A Non I denti c ql Veh i c l e . FJ:e e t ". Oc tober , 313 . Thomas E. Muller , " The Ro l e o f P e r s onal Values Har i sh C . Jain and Rick D . Har i sh C. Mino r i ti e s Novemb e r , . i �}" . ;· �-' J a in , " The Novemb e r 1 9 8 8 . Hacke t t , Pub l i c Po l icy and a Survey " , 315 . •:- - " :> · 1988 . International Tourism Marke t " , 3 14 . '· to Maximi,z e " Empl oyment Equ i ty in Canada : Novemb e r , Recrui tment 1988 . and in C anadian Pol ice Organ i z a t i ons : 1988 . in S egment ing an S e lec t i on of 198 5 - 198 7 " , V i s ib le "' - .-..;;;,_- L f - -- - - �----,--=---- � -